ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846

A+| A| A-

Political Economy of Panchayats in South India

Based on a study of some 500 villages in the four southern Indian states of Andhra Pradesh, Karnataka, Kerala and Tamil Nadu, this paper examines how the functioning of the panchayat system mandated by the 73rd amendment to the Constitution has had an impact on the economic status of villages and the households within them. The study finds that gram panchayats, created by this massive experiment in democratic decentralisation, have had an effect on the delivery of public services, for example, in the targeting of beneficiaries of welfare programmes, but also that positive outcomes are linked to the political elites thrown up by the system.

Political Economy of Panchayats in South India

Based on a study of some 500 villages in the four southern Indian states of Andhra Pradesh, Karnataka, Kerala and Tamil Nadu, this paper examines how the functioning of the panchayat system mandated by the 73rd amendment to the Constitution has had an impact on the economic status of villages and the households within them. The study finds that gram panchayats, created by this massive experiment in democratic decentralisation, have had an effect on the delivery of public services, for example, in the targeting of beneficiaries of welfare programmes, but also that positive outcomes are linked to the political elites thrown up by the system.


he Indian experiment in local village democracy – galvanised by the passage of the 73rd constitutional amendment in 1993 – is among the most ambitious in history. Moreover, the experiment with panchayats in India is of global interest given the array of similar policy experiments going on around the world.1 It comes at a time when economists are re-engaging with political economy issues in their thinking about policy.

It is unsurprising, therefore, to find that panchayats have attracted significant academic attention.2 Since 1993 village governments in India, gram panchayats (GPs), have been responsible for maintaining local amenities such as village roads and drinking water facilities, and for identifying beneficiaries for federal and state poverty alleviation programmes. A key motivation for the 73rd amendment was the belief that local governments may be better placed (than, say, centrally appointed bureaucrats) to identify and respond to villagers’ needs. It was also held that villagers may find it easier to monitor local politicians. Democratisation of the public service delivery system has, thus, been a central element of the Indian decentralisation experiment.

This paper reviews findings from a research project on the political economy of Indian gram panchayats in four states in south India. The research is based on a household and village survey covering 522 villages in Karnataka, Andhra Pradesh, Kerala and Tamil Nadu. The decentralisation experience of these four states shows significant variation [Matthew and Buch 2000]. Karnataka was one of the pioneers of the panchayat movement, and was the first Indian state to mandate regular panchayat elections. Fiscal decentralisation has advanced the most in Kerala, with 40 per cent of state expenditure allocated by panchayats. Kerala panchayats are characterised by high levels of villager participation and regular villager meetings. Andhra Pradesh took a different route and, till recently, sought to energise a political alternative to the panchayat system – the Janmabhoomi programme. Finally, Tamil Nadu continues to have relatively weak gram panchayats with limited devolution of policy powers.

We examine how features of the panchayat system, in particular the design of political institutions, affect how it targets resources towards economically and socially disadvantaged groups in the village. We also examine politician selection. One of the striking features of the panchayat experiment is how it has led to a new political class, many of whom had never held public office previously. We examine how selection mechanisms in Indian villages affect politician outcomes.

The paper is organised as follows. In the first section, we discuss survey design. Section II discusses some background institutional issues. Section III reviews some concrete findings from the research while Section IV pulls together the policy implications.

I Survey Design

Our data come from a village and household level survey conducted in Andhra Pradesh (AP), Karnataka (KA), Kerala (KE) and Tamil Nadu (TN). The survey was conducted during September-November 2002. Sampling occurred in multiple stages, and consisted of purposive sampling up to the level of blocks and random sampling within these blocks. Our final sample consists of 527 villages belonging to 201 elected GPs.3

For each pair of states, we selected two districts (one per state) that shared a common boundary.4 The district pairs were selected, with one exception, to focus on districts that, prior to the linguistic reorganisation of states in 1956, had belonged to the same administrative unit. Our sample consists of nine districts – Bidar (in KA) and Medak (in AP) , Palakkad (KE), Coimbatore (TN), Kasargod (KE), Dakshina Kannada (KA), Dharmapuri (TN), and Chittoor (AP). In KA, we also sampled Kolar district.

For each pair of districts which shared a common boundary three pairs of blocks were selected (that is, three blocks in each of the two districts).5 Two blocks form a pair if they lie in different states but are “linguistically similar”. Using 1991 Census block level data, we defined two blocks to be linguistically similar if, of all the blocks in the district, they have the highest fraction of households with the same mother tongue. The top three matches entered our sample. Linguistic similarity is a good proxy for shared cultural history, given the prevalence of caste and linguistic endogamy. Hence, language matching provides a partial control for “unobservable” socio-cultural differences. The historical and administrative similarity of linguistically matched blocks was checked using princely state maps and the report of the States Reorganisation Commission [GoI 1955].

In AP, KA and TN we randomly sampled six GPs per block and within a GP all villages if the GP had three or fewer villages. If it had more than three villages, then we selected the pradhan’s village and randomly selected two other villages.6 To account for the

Economic and Political Weekly February 24, 2007 much higher GP population in KE we sampled three GPs per block and six wards per GP in KE. This procedure gave a total of 201 GPs and 527 villages. Our survey used four different questionnaires to collect data at the village, politician and household levels.

We conducted household surveys (20 per village) in a random subsample of 259 villages, giving us a sample of 5,180 households. The household questionnaire obtained information on household’s socio-economic status, household structure, views and use of public services in the village, private government benefits. Respondents were also asked to rank problems in the village. Since the sample is divided between male and female and SC/ST and non-SC/ST respondents this provides yet another source of information on gender and caste differences on preferences about village problems.

We also surveyed an elected member of the GP in every village (with precedence given to the GP head if he/she lived in that village) – this gives us a household sample of 544 elected officials. In addition to all the questions on the household questionnaire politicians were also asked a series of questions about their conduct of GP activities.

At the village level we administered a questionnaire using participatory rapid appraisal (PRA) techniques [Chambers 1994] to a group of men selected to represent different caste groups in the village. The PRA questionnaire assessed villager views on problems in the village, the work done by the GP and prevalence of political oligarchy. For the last, we asked respondents to list the extent to which the pradhan, former pradhan and the vice-pradhan controlled prominent activities in the village. We also obtained a detailed listing of castes within the village, and land distribution both within and between castes. The PRA-based questionnaire was separately fielded to (1) a group of women and (2) a group of SC/ST individuals. These PRAs obtained separate measures of women’s and SC/ST problem ranking vis-à-vis public service delivery.

Finally, we undertook an audit of all public goods in the village. This was conducted by an independent enumerator who visually assessed the quality of schools, clinics, roads, drinking water, and sanitation. The enumerator was also asked to identify the extent of GP involvement in improving these facilities.

II Background

It is important to begin by engaging with the details of the panchayat system and its operation. Schedule XI of the Indian Constitution defines the functional items for which states may devolve responsibility to panchayats. While states vary in the extent to which they devolve policy powers to the gram panchayat, most gram panchayats have responsibilities of civic administration in the village together with limited independent taxation powers. On average, roughly 10 per cent of a GP’s total revenue comes from own revenues, with the remainder consisting of transfers from higher levels of government.

While the ambit of GP policy influence varies across Indian states, GPs typically perform (at least) two distinct policy tasks. The first is beneficiary selection for central and state welfare schemes. These are “low spillover” public goods because the benefits are likely to accrue to individual households. These include a variety of transfer programmes such as schemes that provide beneficiary households with funds to acquire housing, private electricity and water supply. Eligibility for these schemes is usually restricted to households below the official poverty line. Most schemes also require that a minimum fraction of beneficiaries be SC/STs.

An important part of a GP’s job, and one on which we focused in our research, is identifying households which are “below the poverty line”, or BPL, households. Possession of a BPL card makes the household eligible for an array of government schemes, ranging from subsidised food through the public distribution system to free hospitalisation. The GP, in collaboration with state government officials, is supposed to identify (via a census) households with income below the poverty line, and prepare the list of BPL households. This list and subsequent selection of beneficiary households under various schemes (from among the BPL households) are supposed to be ratified in gram sabha meetings. All BPL households are eligible for a BPL card, also often provided by the government. This procedure makes the allocation of BPL cards highly political and the success or failure in targeting needy households a key issue. Eighty-seven per cent of the politicians in our sample stated that elected GP politicians were responsible for BPL card allocation.

The second area of GP policy activism is the construction and maintenance of village public goods such as street lights, roads and drains. These are “high spillover” public goods since the benefits accrue more broadly across members of a village. The GP decides the distribution of these public goods within the village and across villages within a GP. It also determines the quality of provision.

Two important institutional features of the gram panchayat which are specific to decentralisation in India are political reservation and village meetings (gram sabhas). The 73rd constitutional amendment mandated political reservation in favour of SC/ST for the pradhan’s7 position, and required that the extent of such reservation in a state reflect the SC/ST population share in that state. It also required that no GP be reserved for the same group for two consecutive elections.

Panchayat legislation also requires that the pradhan consult with villagers and ward members in deciding the choice of beneficiaries and allocation of public goods. This is supposed to be done via village meetings, or gram sabha meetings, called by the elected local government to discuss resource allocation decisions in the village. Seventy-six per cent of the villages in our sample reported having had at least one gram sabha meeting in the last year. However, final decision-making powers in a GP are vested with the pradhan.

III Findings

In this section, we discuss some of the results from the research under three headings. We first study how the panchayat system is targeting resources across households and villages. We then discuss political participation and the operation of gram sabhas. Finally, we will discuss the selection of politicians and its consequences. Table 1 gives the basic data from our sample of households and villages from which the following analyses are drawn.


One of the key problems in targeting public resources is to get them to those who need them most. One key hope for the panchayat system was that it would use the political process to create more effective targeting to needy households and villages. Our data allow us to look at these issues at both levels as we

Economic and Political Weekly February 24, 2007

have detailed data on which transfers households receive and which villages are favoured. We focus on the targeting of scheduled caste and scheduled tribe (SC/ST) households. SC/ST households have suffered from significant historic social and economic disadvantage, leading to worse contemporary outcomes for this group. An important aim of the Indian state’s welfare policy has been to target resources towards this group [Pande 2003].

In Besley et al (2004a), we use our household survey data to measure the provision of household public goods. Here we used only 4,059 households of which 981 were SCs/STs spread across 193 villages8. We measure a household’s exposure to low spillover public goods by a dummy which equals one if it had a house or toilet built under a government scheme or if it received a private water or electricity connection via a government scheme since the last GP election. Approximately 7 per cent of the sample households fall in this category.

We also examine the activity of gram panchayats at the village level. For our key measure – an index of GP activity on high spillover (i e, village level) public goods – we have used information from our audit of village facilities. Specifically, we used an index based on whether the GP undertook any construction or improvement activity on roads, drains, street lights and water sources within a village since the last GP election. (The index is normalised to lie between 0 and 1.) Roughly 79 per cent of our sample villages experienced GP activism on at least one of these public goods.

We were interested in investigating whether the activity at the household or village level is related to reservation status. To capture a village’s reservation status, we constructed an indicator variable equal to one if the village belongs to a GP where the pradhan is reserved for an SC/ST. We used two different variables to measure the political influence of a village – the first equals one if the pradhan resides in that village, and the second equals one if the GP headquarters are in that village.

The results are reported in Table 2. Column (1) reports result from a regression where the left hand side variable is a dummy variable denoting whether the household benefits from a government scheme. The regression includes village fixed effects to account for any factors that differ between villages and affect the extent of targeting. The main finding is that SC/ST households are more likely than non-SC/ST households to receive household transfers, suggesting that these groups are targeted within villages. In column (2) row (2), we examine the interaction between being from an SC/ST group and having a pradhan from a GP that is reserved for an SC/ST. We observe a positive coefficient, indicating that SC/ST households living in a reserved GP are 7 per cent more likely to receive a transfer under government schemes relative to living in non-reserved GPs.

The column (2) regression explores whether this effect is robust to controlling for whether the household lives in the pradhan’s home village. Here we find that there is no benefit to an SC/ ST household from either living in the pradhan’s village or from living in the village where the GP headquarters is located. Thus, it appears the improved targeting of SC/ST households is generated from reservations.

Our results suggest that having a reserved pradhan does indeed further the end of getting better targeting towards SC/ST households. This complements the results on targeting towards women by female pradhans found by Chattopadhyay and Duflo for Rajasthan and West Bengal and the evidence on state level targeting of SC/ST households by SC/ST politicians reported in Pande (2003).

Another important aspect of targeting concerns resource allocation across villages. To examine this, we look at measures of village activism at the village level as measured by the GP

Table 1: Descriptive Statistics


Household data

SC/ST household 0.23

(0.42) Illiterate 0.27

(0.44) Landless 0.38

(0.48) Wealthy 0.29

(0.45) Ever attended gram sabha 0.20

0.40 Possess a BPL card 0.22

(0.43) Received private goods 0.06

(0.24) Politician 0.09


Village data

Literacy rate in 1991 0.42

(0.182) Fraction of villages which had a 0.77 gram sabha in last year (0.41)

Pradhan’s village 0.39

(0.48) GP headquarters 0.31

(0.460) Pradhan reserved for SC/ST 0.20

(0.403) Public good index 0.45


Notes: All variables based on survey data, except the village literacy ratewhich is from the 1991 Census of India. Wealthy is a dummy=1 if household possesses a watch, a fan and either a TV or a radio.

Table 2: Effect of SC/ST Reservation on Resource Allocation

Private Goods Public Goods Household Data Village Data

(1) (2) (3)

SC/ST household 0.048*** 0.041

(0.016) (0.025) SC/ST household* pradhan 0.071** 0.071** Reserved for SC/ST (0.031) (0.031) SC/ST household* pradhan village 0.03

  • (0.025) SC/ST household*GP headquarters -0.019
  • (0.025) Proportion SC/ST households 0.041
  • (0.042) Pradhan village 0.048**

    (0.023) Pradhan reserved for SC/ST -0.003

    (0.039) Pradhan village* pradhan 0.003 Reserved for SC/ST (0.051) GP headquarters 0.041*

    (0.023) Fixed effects Village Village Block Observations 4059 4059 395 R-squared 0.11 0.11 0.67

    Notes: “Private good” is a dummy variable which equals one if the household’s house or toilet was built under a government scheme, or if it received a private water or electricity connection via a government scheme, all since the last GP election. “Public good” is an index of whether the GP undertook any construction or improvement activity on roads, drains, street lights and water sources after the last GP election. The SC/ST household dummy equals 1 for SC/ST households. The pradhan village dummy equals one if the pradhan resides in the given village. The GP headquarters dummy equals 1 if the GP headquarters is located in the village. Robust standard errors in brackets. * significant at 10 per cent; ** significant at 5 per cent; *** significant at 1 per cent.

    Economic and Political Weekly February 24, 2007 activism index. We find that this index is on average 0.05 points higher in the pradhan’s village. This is equivalent to a 10 per cent increase in village activism by panchayats. There is no advantage to being in the GP headquarters. Moreover, having a reserved pradhan has no consequences for village level targeting.

    As Table 1 shows, 39 per cent of our sample villages have pradhans drawn from them. These also tend to be the larger villages within a GP. We interpret the result in column (3) of Table 2 as a consequence of political geography, though we cannot rule out the possibility that larger (and potentially richer) villages both house the pradhan and are more successful at lobbying for public goods. States that have only one village per GP are clearly not going to create an advantage for any particular village whereas those with many villages per GP face a distributional issue. If the pradhan is from a particular village on a repeated basis, then our results suggest that this may create a serious distributional bias.

    Gram Sabhas

    The gram sabha has often been considered the lynchpin of the panchayat system. It has the potential to structure democratic institutions to ensure fair and efficient allocation of public funds. The idea that encouraging citizen participation can improve the workings of a democracy is also echoed in the political science literature. One role for participation emphasised in that literature is to improve the flow of information into the political process beyond that available by electing representatives. Thus, Verba et al (1995:10) characterise political participation as “information rich” acts and observe:

    From the electoral outcome alone, the winning candidate cannotdiscriminate which of dozens of factors, from the position takenon a particular issue to the inept campaign run by the opposition...,was responsible for the electoral victory.

    There are two main ways in which such meetings may improve the workings of government. First, relative to elected representatives, these meetings may better reflect citizens’ preferences on issues such as how to target resources to the most needy groups. Second, by providing a forum for monitoring the actions of elected representatives they may reduce agency problems in politics, and the extent of corruption.

    While holding gram sabhas is compulsory, their frequency and content owes a lot to the discretion of elected officials. Officials from the state or district administration can also have a role in this by choosing not to attend, and therefore making the gram sabha less attractive to hold. It is also the case that a well attended meeting may have no bite on policy-decisions. We exploit our household and village surveys to examine the determinants of participation in gram sabha meetings, and whether having a gram sabha meeting affects beneficiary selection for welfare programmes.

    In our PRA survey, we asked about whether a gram sabha meeting had been held in the past six months. The household survey also asked individuals about gram sabha attendance. Table 1 show that only 77 per cent of villages held gram sabhas in the last year. This suggests a considerable degree of noncompliance with the law. In our household data, Table 1 shows that only around 20 per cent of our sample households report having attended a gram sabha meeting.

    Besley et al (2005a) look at the probability of holding a gram sabha meeting at the village level and find evidence that this is related to village literacy, with more literate villages more likely to hold meetings. Quite why this is true is hard to discern in these data. However, it parallels a larger literature emphasising the civic benefits of greater education.

    In Table 3 based on Besley et al (2005a), we use household data to examine who attends gram sabha meetings, and whether holding gram sabha meetings is correlated with needy households’ access to public welfare as measured by receipt of BPL cards. All specifications include village level fixed effects which control for variation at the village level.

    In column (1) the dependent variable is whether the household respondent attended a gram sabha meeting in the past year. We observe that illiterates are less likely to attend a gram sabha meeting than others. However, this effect is somewhat offset if the household lives in a village with greater overall literacy. This reinforces the idea that literate villages have stronger civic cultures. It is also interesting to observe that SC/ST and landless households are also more likely to attend gram sabha meetings in villages that have a greater number of literate households. Note that this is not a Kerala specific effect since the regressions control for variation at the village level using fixed effects.

    These findings are notable for two reasons. First, there is some suggestion of a political externality from living in a more literate community. Second, gram sabha meetings seem to be a forum used by some of the most disadvantaged groups in the village – the landless and scheduled castes/tribes. This suggests that these groups find the gram sabhas useful and that gram sabha meetings may play some role in moving policy in a direction favoured by these groups. Indeed, a key function of gram sabhas is to target resources to poor households. We now look for evidence of the latter.

    In column (2) we estimate a household regression which exploits within village variation in individual characteristics to examine whether the targeting of BPL cards differs depending on whether the village had a gram sabha in the last year. The results show that illiterate households in villages that have held gram sabhas in the past year are more likely to have a BPL card. There is also evidence of greater targeting of BPL cards towards landless households. In Besley et al (2005a) we interact the characteristics

    Table 3: Gram Sabhas: Participation and Resource Allocation

    Dependent Variable Attended Received
    Gram Sabha BPL Card
    Village Characteristic Literacy Rate Gram Sabha
    (1) (2)
    Illiterate -0.103*** -0.042*
    Illiterate* village characteristic (0.028) 0.183** (0.026) 0.091***
    (0.078) (0.030)
    SC/ST -0.029 (0.040) 0.094** (0.042)
    SC/ST* village characteristic 0.139 0.062
    Landless (0.097) -0.073** (0.047) 0.018
    (0.029) (0.030)
    Landless* village characteristic 0.232*** (0.066) 0.067* (0.035)
    Female -0.086*** -0.009
    Female* village characteristic (0.030) -0.242*** (0.010)
    Upper caste -0.007 (0.018) -0.028* (0.016)
    Wealthy -0.027* -0.079***
    Fixed effects (0.016) Village (0.014) Village
    Observations 5240 5364
    R-squared 0.25 0.4

    Notes: Standard errors in brackets are clustered at village level. Regressions also include respondent age and age squared as controls.

    Economic and Political Weekly February 24, 2007

    that represent disadvantage – illiteracy, landlessness and belonging to a scheduled caste/tribe – with the village literacy rate instead of whether the village had a gram sabha meeting. All three of these interactions are also significant. This does raise the possibility that holding a gram sabha meeting is correlated with other village characteristics that are important in shaping the way in which public resources are targeted. Therefore we cannot say that holding a gram sabha has a causal effect on targeting. This is not an issue we can resolve with the existing data. However, these encouraging results on gram sabhas clearly deserve further careful investigation.

    Our results contribute to a wider debate on how institution design can shape public resource allocation and how the poor can increase their voice in public institutions. It is frequently remarked that poverty is much more than material deprivation and that the poor may have much less voice in the political process. Moreover, a good deal of cynicism attends initiatives to strengthen that voice.

    While the context is very specific, our results sound a more optimistic note. The illiterate, the landless and SC/STs are significantly more likely to attend gram sabha meetings than other groups. Moreover, there appears to be more targeting towards these groups where gram sabha meetings are held. The results are also suggestive of some externalities from literacy in the political process at the village level.

    On a less optimistic note, we find that women are less likely to attend gram sabha meetings than men. Women respondents are around 20 per cent less likely to attend a gram sabha than men. Whether this has significant consequences for public resource allocation need further investigation. But it is clear that the representativeness of gram sabhas is likely to be affected by this. Other tools such as gender reservation in panchayat representation may go some way towards remedying this [Chattopadhyay and Duflo 2004a].

    Political Selection

    By 2000, the 73rd amendment led to the institution of 2,27,698 new village governments, gram panchayats, staffed by over two million elected representatives. Whether these individuals have different skills and motivations from those who previously made political decisions is hard to discern. However, there is a growing body of evidence that political selection is an important consideration in political systems and the current context is an excellent one to think about this.

    The Downsian model of politics, which has dominated political economy for over a generation, has no role for political selection. The role of politics is to seek out the policy position of the median voter, and not to examine who implements that policy. But there is now increasing attention paid to the role of political selection in reforming and improving politics.

    Little is known in general about characteristics of politicians and how they differ from the general population. Table 4 gives some insight into this in the current context. It compares the sample of 540 politicians with the household sample in our data. We report two sets of comparisons. First, we use our entire sample to compare politician and non-politician households, and report the t-test for differences in means across these two populations in Column (3). Second, we restrict attention to SC/ST households and examine whether within this group politician and non-politician households differ (columns (4)-(6)). In all cases, we weight the non-politician sample by the population share of SC/ST households in the village to account for our purposive sampling of SC/ST households.

    Looking across columns (1)-(3), an immediate finding, which is not particularly surprising, is that politicians are an elite group. Politicians, on an average, own five acres of land, which is more than twice the average land ownership of non-politician households. The likelihood that a politician comes from a household with a history of involvement in politics is 25 per cent, as against a mere 6 per cent for non-politician households. Most politicians in our sample do not report politics as their primary occupation. This, in part, reflects the fact that most of them are first-time entrants into politics. While both politician and non-politician households tend to rely on agriculture, politicians are significantly more likely to be cultivators than agricultural labourers. Politicians are also much more likely to be educated, enjoying an average advantage of around three years of education. They are also much more likely to read a newspaper, which reflects greater literacy in the population group.

    Columns (4)-(6) show that reserved politicians who belong to SC/ST also tend to be elites compared to their comparison group (non-politician SC/ST households). Both SC/ST politician and

    Table 4: Selection of Politicians

    Sample: All SC/ST
    Politician Non-politician t-test for Politician Non-politician t-test for
    Difference of Means Difference of Means
    (1) (2) (3) (4) (5) (6)
    Land ownership 5.705 2.025 10.118 2.374 1.055 3.862
    (0.353) (0.086) [0] (0.296) (0.170) [0]
    Family political history 0.258 (0.019) 0.065 (0.005) 9.874 [0] 0.213 (0.037) 0.060 (0.010) 3.900 [0]
    Agricultural labourer 0.066 0.237 -12.660 0.161 0.460 -7.880
    Cultivator (0.010) 0.481 (0.008) 0.217 [0] 11.533 (0.033) 0.315 (0.019) 0.145 [0] 3.830
    (0.021) (0.008) [0] (0.042) (0.014) [0]
    Years of education 7.276 4.750 11.860 6.161 2.880 7.180
    (0.186) (0.103) [0] (0.421) (0.174) [0]
    Newspaper readership 0.692 0.342 15.810 0.613 0.220 8.280
    BPL received (0.019) 0.253 (0.010) 0.230 [0] 1.170 (0.044) 0.363 (0.018) 0.340 [0] 0.460
    (0.018) (0.007) [0.23] (0.043) (0.016) [0.64]
    Public works programme 0.091 (0.012) 0.040 (0.003) 3.904 [0] 0.121 (0.029) 0.084 (0.010) 1.170[0.23]

    Notes: The non-politicians means are weighted by fraction SC/ST households in the village to account for purposive sampling of 4 SC/ST households per village. The p-values for the t-test are provided in square brackets.

    Economic and Political Weekly February 24, 2007 non-politician households are, however, economically disadvantaged relative to the general population. Thus, while reservation is bringing wider participation in politics, it tends to do so by picking elites from among traditionally disadvantaged groups.

    The desirability of having a political elite that is also an economic elite is moot. It may raise concerns about village politics being dominated by a set of narrow interests. However, it may also be the case that the more educated and knowledgeable citizens are better placed to run the village and to provide citizens with the public goods and transfers that they need.

    The last two rows of Table 4 show, however, that the political elite is more likely to claim BPL cards and to participate in public works programmes in spite of its appearing less disadvantaged in economic dimensions. This is evidence of political opportunism. While the difference in unconditional population means for politician and non-politician households receiving a BPL card is insignificant, in Besley et al (2005b) we show that within villages this effect is statistically significant. However, there is evidence that the effect is diminished when politicians are more educated.

    Overall, these results reinforce the need to have a good understanding of the process of political selection and its consequences in village government.

    IV Policy Implications

    The mainstream economics literature now engages with political economy issues in trying to understand what makes government work. The study of panchayats provides an excellent basis for thinking how the insights of modern political economy can contribute to our understanding of public service delivery in India. While the panchayat system has many specific features, there are potentially general lessons for experiments in democratic decentralisation elsewhere.

    The mantra of decentralisation is that it will achieve policies that better reflect the needs of citizens living in village India. Our results on gram sabhas and their link to targeting provide some support for this idea. The fact that reserved politicians target differently in three of our states also reinforces earlier findings that reservation in the gram panchayats can achieve policy change.

    But there are important unresolved issues. Politicians remain opportunistic, undermining the most romanticised view of village government. The fact that this may change with selection of politicians suggests that there needs to be further focus on methods to draft an honest and competent political class. Policy measures that are worth debating include enhanced training for village politicians and use of wage incentives.

    The question of how to ensure that gram sabhas are held is clearly important. Policies here could include better monitoring from above or finding means to enhance the power of citizens to call gram sabhas.

    Finally, our results call for a better debate about political geography and institutional means to guarantee that villages get an equal share of resources, given that the pradhan’s village appears to benefit most of all. This could be in terms of redrawing panchayat boundaries to create more panchayats which contain only few villages or to have a more explicit mechanism for rotating the pradhan’s chair.

    It is clear that there is much yet to be learned about how panchayats work and to think of ways of improving the manner in which they make policy. However, conducting studies based on large samples of villages seems like an important way forward to enhance the quality of policy-making and to understand whether democratic decentralisation is fulfilling its promise.




    1 See Bardhan (2002) and Crook and Manor (1998) for background discussion.

    2 Shortage of space precludes us from surveying the extensive emerging literature. Recent contributions include Bardhan and Mookherjee (2000), Chattopadhyay and Duflo (2004a,b), Foster and Rosenzweig (2001), Chaudhuri and Heller (2004) and Manor (2004).

    3 See Besley et al (2004b) for an extended discussion of the project.

    4 One district in KA (Kolar) that shared boundaries with both AP and TN entered the sample twice. The same holds for one district in AP (Chittoor).

    5 If one district was matched with two different districts then six blocks were chosen from it (three per match). In one block in KE an additional block was sampled as a check on our language matching. This gave us a total of 37 blocks (12 in KA, 9 in AP and TN and 7 in KE).

    6 We excluded all villages with less than 200 persons from our sampling frame. All hamlets with population over 200 were considered as independent villages in drawing the sample.

    7 Also known, depending on the state, as president or sarpanch.

    8 We did not use data from Kerala in this study as the process of allocation in Kerala was very different.


    Bardhan, Pranab (2002): ‘Decentralisation of Government and Development’, Journal of Economic Perspectives, 16(4), pp 185-205.

    Bardhan, Pranab and Dilip Mookherjee (2000): ‘Capture and Governance at Local and National Levels’, American Economic Review, 90(2), pp 135-39.

    Besley, Timothy, Rohini Pande, Lupin Rahman and Vijayendra Rao (2004a): ‘The Politics of Public Good Provision: Evidence from Indian Local Governments’, Journal of the European Economics Association, 2(2-3), pp 416-26.

  • (2004b): ‘Decentralisation in India: A Survey of South Indian Panchayats’, mimeo, London School of Economics.
  • Besley, Timothy, Rohini Pande and Vijayendra Rao (2005a): ‘Participatory Democracy in Action: Survey Evidence from India’, Journal of the European Economics Association.
  • (2005b): ‘Political Selection and the Quality of Government: Evidence from South India’, mimeo. Chambers, Robert (1994): ‘Participatory Rural Appraisal (PRA): Challenges, Potentials and Paradigm’, World Development, 22(10), pp 1437-54.
  • Chattopadhyay, Raghabendra and Esther Duflo (2004a): ‘Women as Policy-Makers: Evidence from an India-wide Randomised Policy Experiment’, Econometrica, 72(5), pp 1409-44.
  • (2004b): ‘Efficiency and Rent-seeking in Local Governments’, Massachusetts Institute of Technology, mimeo.
  • Chaudhuri, Subham and Patrick Heller (2004): ‘Building Local Democracy: Evaluating the Impact of Decentralisation in Kerala’, paper presented to Netsappe III, Paris, June.

    Crook, Richard C and James Manor (1998): Democracy and Decentralisationin South Asia and West Africa: Participation, Accountability, and Performance, Cambridge University Press, Cambridge and New York.

    Foster, Andrew and Mark Rosenzweig (2001): ‘Democratisation, Decentralisation and the Distribution of Local Public Goods in a Poor Rural Economy’, mimeo.

    GoI (1955): Report of the States Reorganisation Commission, Manager of Publications, Government of India, Delhi.

    Manor, James (2004): ‘Democratisation with Inclusion: Political Reforms and People’s Empowerment at the Grassroots’, Journal of Human Development, 5(1), pp 5-29.

    Pande, Rohini (2003): ‘Minority Representation and Policy Choices: The Significance of Legislator Identity’, American Economic Review, 93(4), pp 1132-51.

    Matthew, George and Nirmala Buch (2000): ‘Status of Panchayati Raj in the States and Union Territories of India 2000’, Institute for Social Studies, Delhi.

    Verba, Sidney, Kay Lehman Scholzman and Henry E Brady (1995): Voice and Equality: Civic Voluntarism in American Politics, Harvard University Press, Cambridge Mass.

    Economic and Political Weekly February 24, 2007

    Dear Reader,

    To continue reading, become a subscriber.

    Explore our attractive subscription offers.

    Click here

    Back to Top